Human Centered Technologies for a Safer and Greener European Construction Industry

Projektinformationen

HumanTech

ID Finanzhilfevereinbarung: 101058236

DOI

10.3030/101058236

Projekt abgeschlossen

EK-Unterschriftsdatum 25 Mai 2022

Startdatum 1 Juni 2022

Enddatum 31 Mai 2025

Finanziert unter

• Digital, Industry and Space

Gesamtkosten

€ 9 066 775,89

EU-Beitrag

€ 9 066 775,89

Koordiniert durch

DEUTSCHES FORSCHUNGSZENTRUM FUR KUNSTLICHE INTELLIGENZ GMBH
Germany

CORDIS bietet Links zu öffentlichen Ergebnissen und Veröffentlichungen von HORIZONT-Projekten.

Links zu Ergebnissen und Veröffentlichungen von RP7-Projekten sowie Links zu einigen Typen spezifischer Ergebnisse wie Datensätzen und Software werden dynamisch von OpenAIRE abgerufen.

Leistungen

Demonstrators, pilots, prototypes (6)

Pilot II - Human-Robot collaboration and wearables

Pilot implementation of Human Robot Collaboration scenario and wearables with full description and evaluation. This deliverable is connected to Task T7.2

Pilot I - Dynamic semantic digital twin

Pilot implementation of Dynamic semantic digital twin generation with full description and evaluation. This deliverable is connected to Task T7.1

Pilot IV - Bridge inspection and monitoring

Pilot implementation of UAV data acquisition and information extraction with link to BIMxD with full description and evaluation. This deliverable is connected to Task T7.4

Pilot V - Robotic mastic application

Pilot implementation of robotic mastic application with BIMxD task planning, full description and evaluation. This deliverable is connected to Task T7.5

Pilot III - Remote controlled demolition

Pilot implementation of remote controlled demolition with HumanTech robotic platform, BIMxD demolition ontology generation and evaluation. This deliverable is connected to Task T7.3

Mobile platform demonstration

Demonstration of the developed mobile platform for construction applications. This deliverable is connected to Task T5.3

Documents, reports (24)

Complete Scan-to-BIMxD Digital Twin pipeline

Description of the full pipeline with detailed process charts and software prototypes. This deliverable is connected to Task T3.6

Body sensor network with integrated camera approach

The deliverable is a report describing the hardware setup and characteristics of the visual-inertial body sensor network, including the sensor locations on the body and their fixations, the synchronisation mechanism, the spatial registration approach (registration between camera and IMU network, and their registration to the body model), the fusion approach (how camera and IMU measurements are used for kinematics estimation). It also documents the body model with its coordinate systems and, if applicable, anatomical landmarks. This deliverable is connected to Task T4.1

Scientific report on unseen object class and shape estimation for robotic grasping

Description of the developed machine learning algorithms for classification, shape and pose estimation of unseen objects intended for autonomous robotic grasping of construction material. This deliverable is connected to Task T5.4

Scientific report on teaching control policies by teleoperation

"Description of the methods of ""teaching by demonstration"" implemented through teleoperation. The demonstrations carried out by the human operator will be used to extract control ""policies"", which will be described here. This deliverable is connected to Task T5.5"

HT worker assessment report - final

This deliverable details the evaluation and the experimentation procedure focused on the HT wearables system and human-robot interactions together with the conclusions and potential improvements in needed. This deliverable is connected to Task T6.2

Workflow capturing scientific report

Report describing a method of automatic capturing and machine understanding of a workflow based on dynamic semantic digital twin approach of HumanTech. This deliverable is connected to Task T6.3

Scientific report on hybrid haptic-visual feedback mechanisms for efficient teleoperation of demolition robots

Report on developments and experiments to improve dexterity during teleoperation using haptics. This deliverable is connected to Task T5.2

BIMxD IDM with classes and attributes according to ISO 29481 and respective standards

BIMxD IDM with classes and attributes according to ISO 29481 and respective standards. t will be developed a PM and related exchange requirements for expanding existing BIM representations with such for robots, human-robotic task planning, worker pose, and action, state of tasks, starting from existing references.Templates for open source BIMxD authoring. This deliverable is connected to Task T2.1

Wearable-user localization algorithm

A report describing the user localization algorithm approach using image data acquired from the chest-worn wide angle camera. This deliverable is connected to Task T4.3

HT micro-learning unit descriptors

Short description of each of the Learning Units and their learning outcomes, delivery methodologies, content and available resources. This deliverable is connected to Task T6.1

bSDD specification from BIMxD

A set of bSDD specifications will be provided, after identifying classifications and their properties, allowed values, units and translations with respect to HT innovation of the project. The BIMxD specifications will be used to develop the bSDD specifications. This deliverable is connected to Task 2.4

HT safety, gender and ethics report

Report summarizing the findings from the iterative user evaluations of the Human-Tech technologies as well as the according ethical procedures. This deliverable is connected to Task T6.4

HT worker assessment report - initial

This deliverable provides an initial version of the methods for evaluating usability, trust in automation and discomfort, including the procedure for the lab-based experimentation. This deliverable is connected to Task T6.2

The HT vision and research requirements

A qualified report stating the vision for all HumanTech partners and identified research requirements. D1.1 is connected to Tasks T1.1, T1.2

Communication framework for control and supervision

Description of communication framework used for supervision and control of physical robot and the on-site or virtual environment. The description includes the design and the actual implementation. This deliverable is connected to Task T5.6

HT user requirements and ethics approach

Specification of user requirements and description of the ethical approach for data collection for user studies as part of the HumanTech framework. This deliverable is connected to Task T1.4

Training monitoring and evaluation report

Reporting on the monitoring and evaluation results from each delivery of HT Training Learning Units. This deliverable is connected to Task T6.1

Intelligent exoskeleton with intention prediction prototype

"A report on the design and validation of mechanisms to render exoskeletons easier to use in polyvalent work environments under the concept of ""Intelligent Transparency"". This deliverable is connected to Task T4.2"

IFC schema extension

After an in-depth study of the existing IFC schema (classes, Predefined Type, relations and attributes) an IFC extension will be suggested from BIMxD specifications and transferred into standardization processes, Enhancing a discussion with the buildingSMART technical and product room. This deliverable is connected to Task T2.5

HT Impact Master Plan

Document detailing the project dissemination, exploitation and communication plans, outlining the target groups and their segments. This deliverable is connected to Task T8.1

HT Dissemination and Communication Interim Report

Documentation of the actions and outcomes achieved concerning the promotional strategies (communication & dissemination) during the first project period. KPIs and mitigation measures will be presented. This deliverable is connected to Task T8.2

HT Dissemination and Communication Final Report

Documentation of the actions and outcomes achieved concerning the promotional strategies (communication & dissemination) during the second project period. KPIs and mitigation measures will be presented. This deliverable is connected to Task T8.2

Hyperspectral material scanning approach

The deliverables will include a sensor rig model with a bill of materials which will include the list of sensors required to perform hyperspectral material scanning. Architecture diagrams for sensor integration, drivers for sensor fusion, and a detection software module for hyperspectral material scanning that is intended to work with the target sensor suite. This deliverable is connected to Task T3.3

Development of the ontology for demolition task planning

This deliverable is a report describing the developed ontology for the automatic execution of demolition activities. It will include a detailed list of the classes, attributes, relations, and all relevant components needed to provide knowledge about building demolition and to represent the demolition environment. This deliverable is connected to Task T5.1

Other (3)

BIMxD XR visualization prototype

XR-glass integration and BIMxD visualization framework (HL, HC, SCT, DFKI, STAM, CAT) M16-M28: Integration of an XR-glass such as the Microsoft Hololens. Development of a calibration approach between wearable camera-based localization and local map reconstructed by XR-glass. Integration of remote processing connectivity for visualization of high-quality localized BIMxD models on-site. This deliverable is connected to Task T4.4

Open source BIM authoring tools: Framework and Software Modules

An open source framework for editing and updating BIMxD data representation based on IFC. This deliverable is connected to Task T2.2

3D point cloud semantic segmentation algorithm

Description of developed machine learning algorithm for semantic segmentation of 3D point cloud building data. This deliverable is connected to Task T3.5

Data sets, microdata, etc (1)

Domain-adapted synthetic dataset for 3D semantic segmentation

A synthetic dataset for 3D Semantic Segmentation in buildings, transformed by machine learning methods developed within HumanTech to increase realism. This deliverable is connected to Task T3.4

Veröffentlichungen

Peer reviewed articles (5)

Federating cross-domain BIM-based knowledge graph

Autoren: Wojciech Teclaw, James O’Donnel, Ville Kukkonen, Pieter Pauwels, Nathalie Labonnote, Eilif Hjelseth
Veröffentlicht in: Advanced Engineering Informatics, Ausgabe 62, 2024, ISSN 1474-0346
Herausgeber: Elsevier BV
DOI: 10.1016/J.AEI.2024.102770

Robot Learning from Teleoperated Demonstrations: A Pilot Study Towards Automating Mastic Deposition in Construction Sites

Autoren: Irati Rasines, Erlantz Loizaga, Rebecca Erlebach, Anurag Bansal, Sara Sillaurren, Patricia Rosen, Sascha Wischniewski, Arantxa Renteria, Itziar Cabanes
Veröffentlicht in: Robotics, Ausgabe 14, 2025, ISSN 2218-6581
Herausgeber: MDPI AG
DOI: 10.3390/ROBOTICS14080114

OPA-3D: Occlusion-Aware Pixel-Wise Aggregation for Monocular 3D Object Detection

Autoren: Yongzhi Su , Yan Di , Guangyao Zhai , Graduate Student Member, IEEE, Fabian Manhardt, Jason Rambach, Benjamin Busam , Didier Stricker, and Federico Tombari
Veröffentlicht in: IEEE Robotics and Automation Letters, 2023, ISSN 2377-3766
Herausgeber: IEEE ROBOTICS AND AUTOMATION LETTERS, VOL. 8, NO. 3, MARCH 2023
DOI: 10.1109/LRA.2023.3238137

Resolving Symmetry Ambiguity in Correspondence-Based Methods for Instance-Level Object Pose Estimation

Autoren: Yongliang Lin, Yongzhi Su, Sandeep Inuganti, Yan Di, Naeem Ajilforoushan, Hanqing Yang, Yu Zhang, Jason Rambach
Veröffentlicht in: IEEE Transactions on Image Processing, Ausgabe 34, 2025, ISSN 1057-7149
Herausgeber: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/TIP.2025.3544142

Robots Adapting to the Environment: A Review on the Fusion of Dynamic Movement Primitives and Artificial Potential Fields

Autoren: Irati Rasines, Itziar Cabanes, Anthony Remazeilles, Joseph Mcintyre
Veröffentlicht in: IEEE Access, Ausgabe 12, 2024, ISSN 2169-3536
Herausgeber: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/ACCESS.2024.3418573

Conference proceedings (12)

HiPose: Hierarchical Binary Surface Encoding and Correspondence Pruning for RGB-D 6DoF Object Pose Estimation

Autoren: Yongliang Lin; Yongzhi Su; Praveen Nathan; Sandeep Inuganti; Yan Di; Martin Sundermeyer
Veröffentlicht in: 2024
Herausgeber: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI: 10.1109/CVPR52733.2024.00967

Task Automation in Construction Sites: Robot Learning from Teleoperated Demonstrations

Autoren: Irati Rasines, Miguel Prada, Alfonso Dominguez, Elisa Prada, Anthony Remazeilles, Itziar Cabanes Axpe
Veröffentlicht in: 2024 7th Iberian Robotics Conference (ROBOT), 2024
Herausgeber: IEEE
DOI: 10.1109/ROBOT61475.2024.10796867

Annotation rules and classes for semantic segmentation of point clouds for digitalization of existing bridge structures

Autoren: Marius Schellen, Fabian Kaufmann , Christian Glock and Thomas Tschickardt
Veröffentlicht in: 2023 European Conference on Computing in Construction, 2023
Herausgeber: 2023 European Conference on Computing in Construction
DOI: 10.35490/EC3.2023.228

U-RED: Unsupervised 3D Shape Retrieval and Deformation for Partial Point Clouds

Autoren: Yan Di , Chenyangguang Zhang , Ruida Zhang , Fabian Manhardt , Yongzhi Su , Jason Rambach , Didier Stricker , Xiangyang Ji and Federico Tombari
Veröffentlicht in: ICCV Conference 2023, 2023
Herausgeber: ICCV Conference 2023
DOI: 10.48550/arXiv.2308.06383

BRep Boundary and Junction Detection for CAD Reverse Engineering

Autoren: Sk Aziz Ali, Mohammad Sadil Khan, Didier Stricker
Veröffentlicht in: 2024 IEEE 3rd International Conference on Computing and Machine Intelligence (ICMI), 2024
Herausgeber: IEEE
DOI: 10.1109/ICMI60790.2024.10585950

IFC Properties Validation Using Deep Graph Neural Network

Autoren: Wojciech Teclaw, Reidar Kind, Nathalie Labonnote
Veröffentlicht in: 2024 9th International Conference on Smart and Sustainable Technologies (SpliTech), 2024
Herausgeber: IEEE
DOI: 10.23919/SPLITECH61897.2024.10612441

ToF-360 – A Panoramic Time-of-flight RGB-D Dataset for Single Capture Indoor Semantic 3D Reconstruction

Autoren: Hideaki Kanayama; Mahdi Chamseddine; Suresh Guttikonda; So Okumura; Soichiro Yokota; Didier Stricker; Jason Raphael Rambach
Veröffentlicht in: 2025
Herausgeber: IEEE/CVF (Hrsg.). 21st. CVPR Workshop on Perception Beyond the Visible Spectrum (PBVS-2025), June 11-15, Nashville, Tennessee, USA, IEEE, 2025.
DOI: 10.57967/HF/5074

BIM-based construction quality assessment using Graph Neural Networks

Autoren: Navid Kayhani, Brenda McCabe, and Bharath Sankaran
Veröffentlicht in: 2023 Proceedings of the 40th ISARC, Chennai, India, 2023
Herausgeber: 2023 Proceedings of the 40th ISARC, Chennai, India
DOI: 10.22260/ISARC2023/0004

SG-PGM: Partial Graph Matching Network with Semantic Geometric Fusion for 3D Scene Graph Alignment and its Downstream Tasks

Autoren: Yaxu Xie, Alain Pagani, Didier Stricker
Veröffentlicht in: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Herausgeber: IEEE
DOI: 10.1109/CVPR52733.2024.02683

Ontology-Based Semantic Labelling for RGB-D and Point Cloud Datasets

Autoren: Fabian Kaufmann, Mahdi Chamseddine, Suresh Guttikonda, Christian Glock, Didier Stricker and Jason Rambach
Veröffentlicht in: 2023 European Conference on Computing in Construction – 40th International CIB W78 Conference, 2023
Herausgeber: 2023 European Conference on Computing in Construction – 40th International CIB W78 Conference
DOI: 10.35490/EC3.2023.241

Single Frame Semantic Segmentation Using Multi-Modal Spherical Images

Autoren: Suresh Guttikonda, Jason Rambach
Veröffentlicht in: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2024), 2023
Herausgeber: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2024)
DOI: 10.48550/arXiv.2308.09369

How many Robots is too many? Findings about Single-Human Multiple-Robot Systems

Autoren: Estefany Rey-becerra, Sascha Wischniewski
Veröffentlicht in: AHFE International, Human Interaction and Emerging Technologies (IHIET 2024), Ausgabe 157, 2024
Herausgeber: AHFE International
DOI: 10.54941/AHFE1005460